A silicon wafer is etched by using different photographic masks to produce large-scale integrated (LSI) semiconductor circuits, very-large-scale integrated (VLSI) semiconductor circuits or ultra-large scale integrated (ULSI) semiconductor circuits. In general, a photographic mask is imprinted with a predetermined pattern. The pattern is then imprinted on the wafer by various lithographic methods. It shall be appreciated by those skilled in the art that the pattern imprinted on the wafer should be essentially identical to the predetermined pattern. Any deviation from the predetermined pattern would constitute a defect and shall render that wafer defective. Also, any imperfection on the non-etched surface of the silicon wafer (such as pits, scratches, foreign matter particles etc.) may also constitute an undesirable defect.
Hence, it is important to detect such defects. When a defect is detected, either at least a portion of the wafer is discarded or the defect is analyzed to determine if it constitutes a critical defect or a nuisance defect (a defect which will not adversely affect performance). Therefore it is important that the inspection system does not declare “false defects” (i.e., declare a defect for a non-defective wafer).
Semiconductor circuits are becoming more and more complex, condensed in structure and smaller in size, and are thus more prone to defects. A conventional method for detecting defects often includes several different test procedures.
Conventional optical inspection system, utilize bright-field, dark-field and gray-field detection based techniques. Bright-field, dark-field and gray-field based techniques, are generally defined as imaging techniques wherein the detected image is completely bright, completely dark, or partially bright, respectively, in the absence of a specimen.
In a simple bright-field based technique, an illumination system illuminates a specimen from above, and a collection optical system located above or below the specimen, detects the light reflected or scattered from the specimen. The broadest definition of bright-field light refers to light thus collected. It is noted, however, that other techniques use different definitions of a bright-field light beam, as shall be described herein below.
In a typical dark-field based technique, either the specimen is illuminated from above and light reflected there from is collected from the sides, or the specimen is illuminated from the side and light reflected there from is collected from above. The light thus collected is typically referred to as the dark-field light beam. A typical gray-field based technique shall be discussed herein below with reference to FIG. 1.
U.S. Pat. No. 6,178,257 entitled “Substrate Inspection Method and Apparatus”, U.S. Pat. No. 5,699,447 entitled “Two-Phase Optical Inspection Method and Apparatus for Defect Detection”, and U.S. Pat. No. 5,982,921 entitled “Optical inspection method and apparatus”, all issued to Alumot et al. and assigned to the assignee of the present disclosed technique, describe a dark-field detection system, and are incorporated herein by reference. U.S. Pat. No. 6,122,046 issued to Almogy et al., entitled “Dual Resolution Combined Laser Spot Scanning and Area Imaging Inspection”, and assigned to the assignee of the present disclosed technique, describes a detection system utilizing a technique combining dark-field and bright-field imaging, and is also incorporated herein by reference.
U.S. Pat. No. 6,259,093 issued to Wakiyama et al., entitled “Surface Analyzing Apparatus”, is directed to an apparatus for the detection of foreign matter and defects on a wafer surface. A polarized laser light is scattered from the wafer surface and is detected in an optical microscope. Since the pattern imprinted on the wafer causes a constant polarization in the reflected light, an appropriate polarizing mask on the microscope side, reduces the intensity of the reflected light. However, the light reflected from surface defects and foreign matter is significantly less influenced by the mask and hence does not exhibit a reduction in brightness, and is therefore detectable (i.e., distinguished from the polarized light).
U.S. Pat. No. 5,699,447 issued to Alumot et al., entitled “Two-phase optical inspection method and apparatus for defect detection”, is directed to a method for detecting defects on patterned wafers. The wafer is illuminated and light diffracted from the wafer surface is collected by a plurality of detectors, arranged in a circular pattern around the inspected wafer.
U.S. Pat. No. 6,064,517 issued to Chuang et al., entitled “High NA System for Multiple Mode Imaging”, is directed to an inspection apparatus which provides different imaging modes, such as dark-field and bright-field. The apparatus includes a high numerical aperture catadioptric (using both reflection and refraction to form an image) optical group which forms an intermediate image, the image is then corrected for aberrations by a focusing group and mapped to a plane located at a pupil of the system. Apertures placed at this plane can be used to limit the range of scattering angles reaching the image detector.
U.S. Pat. No. 6,122,046 issued to Almogy et al., entitled “Dual Resolution Combined Laser Spot Scanning and Area Imaging Inspection”, is directed to an apparatus for optically detecting defects in a silicon substrate. A linearly polarized light beam is passed through a beam splitter, which is aligned so as to transmit the illuminating light beam without deflection. The illuminating light beam then passes through a quarter wave plate which circularly polarizes it. The illuminating light beam is then reflected from the inspected surface. The reflected light passes through the quarter wave plate in the opposite direction and is linearly polarized thereby, but in a direction perpendicular to the original linear polarization direction of the illuminating light beam. The reflected light beam is deflected by the beam splitter, due to its perpendicular polarization, toward a bright-field detector.
The article “Detection of Fibers by Light Diffraction”, J. List et al. (1998), describes an apparatus for the detection of asbestos fibers in air flow. The device described detects light scattered from the fibers in the air. A pulsed Nd:YAG laser produces a high intensity illuminating light beam. An apertured mirror is located at the opposite side of the laser source, admitting the illuminating light beam toward a light trap and deflecting light, scattered by the asbestos fibers in the air, toward a light detector (CCD). The apertured mirror protects the light detector from the high intensity illuminating light beam.
Reference is now made to FIG. 1, which is a schematic illustration of a system, generally referenced 10, for scanning a wafer surface, which is known in the art. System 10 is used for scanning a wafer surface 12. System 10 includes a laser light source 14, a scanner 16, a polarizing beam splitter 20, a quarter wave plate 24, an objective lens assembly 26, a relay lens assembly 32, an annular mirror 34, a bright-field detector 36 and a gray-field detector 38.
Laser light source 14, scanner 16, polarizing beam splitter 20, quarter wave plate 24 and objective lens assembly 26 are positioned along a first optical axis 60. Polarizing beam splitter 20, relay lens assembly 32, annular mirror 34 and bright-field detector 36 are positioned along a second optical axis 62. Annular mirror 34 and gray-field detector 38 are positioned along a third optical axis 64.
Polarizing beam splitter 20 includes a semi-transparent reflection plane 22. Reflection plane 22 is oriented at 45 degrees relative to wafer surface 12. Annular mirror 34 is oriented at 45 degrees relative to optical axes 62 and 64. For purposes of simplicity, objective lens assembly 26 is depicted in 2A as a basic objective lens assembly, including an aperture stop 28, located at a pupil of the scanning system, and an objective lens 18. Objective lens 18 has a focal length F.sub.1. Aperture stop 28 has a diameter D.sub.P.
Laser light source 14 emits a laser light beam 40, which is then received by scanner 16. Scanner 16 expands and redirects laser light beam 40, thereby emitting alternating illuminating light beams at dynamically changing angles and a constant diameter D. The example illustrated in FIG. 1 shows only an illuminating light beam 44, having a maximal scanning angle .theta., relative to optical axis 60. Other illuminating light beams (not shown) have scanning angles between theta. and -.theta.
Illuminating light beam 44 passes through polarizing beam splitter 20 and from there, further through quarter wave plate 24. Quarter wave plate 24 circularly polarizes illuminating light beam 44 in a first angular direction. Illuminating light beam 44 enters objective lens assembly 26 and passes through aperture stop 28. Objective lens assembly 26 focuses illuminating light beam 44 onto a point 30.sub.1 on wafer surface 12.
Illuminating light beam 44 is reflected and scattered from point 30.sub.1, in a plurality of directions. Some of the reflected and scattered light is collected by the objective lens assembly, and used to detect the properties of wafer surface 12. According to this technique, the collected light includes a bright-field light beam portion and a gray-field light beam portion. The bright-field light beam is defined as light the portion of collected light which follows the exact path of the illuminating light beam. The gray-field light beam is defined as the rest of the collected light, which is not included in the bright-field light beam. A bright-field light beam 50 and a gray-field light beam 52, of the light scattered and reflected from point 30.sub.1, are collected by objective lens 18. Objective lens 18 collimates bright-field light beam 50 and gray-field light beam 52, and directs the light beams through aperture stop 28.
Light beams 50 and 52 exit from objective lens assembly 26 circularly polarized in the opposite angular direction as illuminating light beam 44. Light beams 50 and 52 pass through quarter wave plate 24, and become linearly polarized, perpendicular to the polarization of illuminating light beam 44. Light beams 50 and 52 are then reflected off semi-transparent reflection plane 22, and directed to relay lens assembly 32.
Relay lens assembly 32 produces an inverted image of the pupil of aperture stop 28, at the pupil of annular mirror 34. Bright-field light beam 50 passes through the aperture of annular mirror 34. Bright-field detector 36 receives bright-field light beam 50 and detects the intensity thereof. Gray-field light beam 52 is reflected off annular mirror 34. Gray-field detector 38 receives gray-field light beam 52 and detects the intensity thereof.
It is noted that the light beams emitted at other times and having other scanning angles, reach other points on wafer surface 12, between point 30.sub.1 and another point 30.sub.2, which is located on the opposite side of optical axis 60 from point 30.sub.1.
System 10 further includes additional objective lens assemblies (not shown), which are interchangeable with objective lens assembly 26. These objective lens assemblies are mounted on a turret, a slide (both not shown), and the like, which enables interchanging objectives. Each of the different objective lens assemblies is used for a different mode of operation.
Reference is further made to FIGS. 2A and 2B. FIG. 2A is a schematic illustration of objective lens assembly 26 and scanned wafer surface 12 of system 10 (FIG. 1). FIG. 2B is a schematic illustration of an additional objective lens assembly 102 which replaces objective lens assembly 26 (FIG. 2A), and scanned wafer surface 12.
With reference to FIG. 2B, objective lens assembly 102 replaces objective lens assembly 26 (FIG. 2A). It is noted that the optical elements of objective lens assembly 102 are not shown. Objective lens assembly 102 has a focal length F.sub.2 equal to 1 1 2 F 1 wherein F.sub.1 is the focal length of objective lens assembly 26 (FIG. 2A).
Objective lens assembly 102 receives an illuminating light beam 110.sub.1 of diameter D, and focuses it onto a point 120.sub.1 on wafer surface 12. Illuminating light beam 110.sub.1 is similar to illuminating light beam 44 (FIG. 2A), having a maximal scanning angle theta.
Objective lens assembly 102 collects a bright-field light beam 110.sub.1 having diameter D and a gray-field light-beam 112.sub.1 having diameter D.sub.P. It is noted that objective lens assembly 102 may include various optical elements (e.g., lenses, stops, and the like), which are not shown.
The line between points 30.sub.1 and 30.sub.2 (FIG. 2A) on wafer surface 12, is known as the scan line of system 10. It is well known that the scan line length is proportional to the focal length of the objective lens assembly. Hence, the scan line length for the system of FIG. 2B, is approximately ½ of the scan line length of system of FIG. 2A. Furthermore, it is well known that the scanning speed of a scanning system such as system 10 (FIG. 1), is proportional to the square of the scan line length.
It is also well known that the numerical aperture of the scanning light beams of system 10 is inversely proportional to the focal length of the objective lens assembly used. Hence, the numerical aperture for system of FIG. 2B, is approximately 2 times the numerical aperture for system of FIG. 2A. Furthermore, is well known that the scanning resolution for a scanning system such as system 10 is proportional to the numerical aperture.
Thus, by selecting different objectives with different focal lengths, the user of system 10 can choose between a low-speed, high-resolution and a high-speed, low-resolution scan. It is noted that to increase the gray-field numerical aperture, it is required to increase both the numerical aperture of the objective lens assembly and the size of the polarizing beam splitter. The cost of an objective lens assembly and the cost of a polarizing beam splitter, are highly correlated with their respective sizes. Hence, increasing the gray-field numerical aperture for system 10 involves a significant cost increase. It is still further noted that the objective lens assembly is the element of system 10 which is closest to the wafer, located directly there above. Hence, replacing objectives when changing magnification modes, involves a risk of contaminating the inspected wafer.